Author Topic: PSU floating ground  (Read 2544 times)

0 Members and 1 Guest are viewing this topic.

Offline bitmanTopic starter

  • Supporter
  • ****
  • Posts: 299
  • Country: us
  • Open Source Guy jabbing with Electronics
PSU floating ground
« on: May 06, 2017, 07:47:22 pm »
Question: If floating ground is bad, why does (my) DC PSU have both ground and 0 volts? Why wouldn't the PSU come with ground and 0 volt already connected into a single jack?

What I'm trying to understand is when connecting 0 to ground is BAD when it comes to DC voltage. I understand the benefits of connecting them, but I'm puzzled why you wouldn't always do that.

Issue: My PSU has both ground and 0 volt. However, there's no easy connect to ground to 0 volt without impacting the bananaplug females. In particular my BNC plugs are only wide enough to go from + to GND - not from + to 0 volt. Of course if I connect 0 to GND I can use the GND connection with the + volt for the BNC connector.  What I really need to understand is why I wouldn't just make a permanent connection between GND and 0 volt and be done with it.
 

Offline retrolefty

  • Super Contributor
  • ***
  • Posts: 1648
  • Country: us
  • measurement changes behavior
Re: PSU floating ground
« Reply #1 on: May 06, 2017, 08:33:23 pm »
Let me try and give a simple example why you would not want to always ground the "0" terminal on a DC supply.

 If you had a circuit that required both a positive +15vdc and negative -15vdc voltages. In that cased the proper 'grounding' would be ground to positive terminal on one supply (the negative supply) and ground to 0 terminal of the other supply (the positive supply).

 You would then use three wires to your circuit, -15, +15, and ground. For most usefulness one would not want to be forced to have either terminal of a DC power supply always grounded, but rather be able to 'jumper' either terminal to ground as the circuit under test requires.

 

Offline bitmanTopic starter

  • Supporter
  • ****
  • Posts: 299
  • Country: us
  • Open Source Guy jabbing with Electronics
Re: PSU floating ground
« Reply #2 on: May 06, 2017, 08:48:38 pm »
Let me try and give a simple example why you would not want to always ground the "0" terminal on a DC supply.

 If you had a circuit that required both a positive +15vdc and negative -15vdc voltages. In that cased the proper 'grounding' would be ground to positive terminal on one supply (the negative supply) and ground to 0 terminal of the other supply (the positive supply).
Thanks. I knew that, however what I failed to mention is that my cheap PSU only has one + and one - terminal. Ie. I cannot connect them together to achieve the dual power scenario here.  Given my basic setup, for now I'm running with the two always connected and I cannot think of any scenario where I would need to disconnect them?  My PSU is so basic, it doesn't even provide AC :)
 

Offline retrolefty

  • Super Contributor
  • ***
  • Posts: 1648
  • Country: us
  • measurement changes behavior
Re: PSU floating ground
« Reply #3 on: May 06, 2017, 09:49:38 pm »
I suspect that you are making it more confusing then it needs to be. How about this: Any circuit being powered requires two connections to operate, a 2 wire voltage source. There is no need, and sometimes required, that neither wire to be 'grounded', which then is named an "isolated voltage".

 There are situations where one must ground one of the two power supply terminals, but that is determined by the circuit under test and whatever else might be wired to it. Best supplies offer a choice of isolated or ground referenced voltage, and having the option is always better off. Recall that battery powered devices require no ground connection to operate, as a battery is an example of a isolated voltage source.

 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 19522
  • Country: gb
  • 0999
Re: PSU floating ground
« Reply #4 on: May 06, 2017, 10:52:16 pm »
Let me try and give a simple example why you would not want to always ground the "0" terminal on a DC supply.

 If you had a circuit that required both a positive +15vdc and negative -15vdc voltages. In that cased the proper 'grounding' would be ground to positive terminal on one supply (the negative supply) and ground to 0 terminal of the other supply (the positive supply).
Thanks. I knew that, however what I failed to mention is that my cheap PSU only has one + and one - terminal. Ie. I cannot connect them together to achieve the dual power scenario here.  Given my basic setup, for now I'm running with the two always connected and I cannot think of any scenario where I would need to disconnect them?  My PSU is so basic, it doesn't even provide AC :)
One problem with a power supply with its negative rail earthed referenced is, if you were to connect it to another device which floats at a different voltage, with respect to earth. The result would be a short circuit.

If all the power supply is doing is powering one device and there are no other power supplies involved, then it really makes no difference whether it's earthed referenced or not.
 
The following users thanked this post: bitman

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16615
  • Country: us
  • DavidH
Re: PSU floating ground
« Reply #5 on: May 07, 2017, 01:42:07 am »
Say your power supply internally connects its minus output to ground.  Now when you connect your oscilloscope ground to the minus side at the device under test, the voltage drop across the minus lead is now connected to ground through the power supply *and* through the oscilloscope resulting in a ground loop and some of the power supply current flows through the ground loop including the oscilloscope and not the minus lead to the device under test.

Except for noise considerations, that is not such a big deal but say the power supply negative lead falls off and you do not notice.  Now *all* of the power supply current goes through the oscilloscope ground lead.  That may not be such a big deal either but say now the device under test shorts out; the resulting current surge may be 10s or 100s of amps resulting in such things as the oscilloscope's probe ground turning white hot and melting as happened to me once or damage to the oscilloscope.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf